WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: Screen reader interpretations of images in text (not part of a link)

for

From: Steve Faulkner
Date: May 8, 2012 8:30AM

hi jared,
you wrote:

"It really is a double-edged sword. If you flag them as errors, it may
promote accessibility. Or it may simply encourage authors to add null
or bogus alt attribute values (which suggest an image alternative
where none is present) simply to generate valid HTML. Admittedly, this
is the current situation for HTML <5, so I wouldn't really expect
things to change much."



if this were the case it would be expected that there would be a
correlation between high number of alt="" and low HTML errors or conforming
pages.
I have done some ad hock data collection from the top 10000 web sites data
[1] i collected recently and have not found any correlation so far.


"Allowing the absence of the alt attribute altogether could be
absolutely clear - the image is inaccessible and does not have a known
alternative. There's no ambiguity or burden on the user to
differentiate between true alternative text and something else that is
currently in the same place the alternative would go if it were known."


The idea that an img missing an alt can equal an informative image for
which there is no text alternative provided would only work IF the page was
other wise conforming HTML5. This does not appear to be the case.


Note that figcaption is different as the figcaption element has a defined
semantic (IA2 role of caption as implemented in firefox)) and the figure
element will be getting a role in ARIA 1.1 (cy=urrently implemented as the
more geenric group role in firefox). what this means in the case of an
image missing an alt is that if its in a figure you can be fairly certain
it is a meaningful image. The caption is the label for the figure and SRs
should announce the presence of the image even though it does not have an
alt (becuase its inside a figure element)

<figure>
<img>
<figcaption>image 1<figcaption>
</figure>

i go into more details about this in HTML5 Accessibility Chops: the figure
and figcaption elements [2]

[1] http://www.paciellogroup.com/blog/?sÚta+for+the+masses
[2]
http://www.paciellogroup.com/blog/2011/08/html5-accessibility-chops-the-figure-and-figcaption-elements/


regards
steve

On 7 May 2012 20:29, Jared Smith < <EMAIL REMOVED> > wrote:

> On Mon, May 7, 2012 at 12:34 PM, Steve Faulkner wrote:
>
> Thanks for chiming in Steve. I knew if I was wrong that you'd set me
> straight.
>
> > success or failure techniques are not normative [1]
>
> Of course. Though the intention is that if a page has a failure, it
> would rarely be conformant. In this case, assuming HTML5 remains as
> is, it seems the failure could simply be updated to not require the
> alt attribute if the alternative text is provided elsewhere.
>
> > "In HTML5, the alt attribute is currently optional."
> >
> > I think it is more correct to say that it may be omitted in one
> > circumstance: when a programmatically associated text alternative is
> > provided using the figure/figcaption elements. [4]
> >
> > There are no other circumstances where it may be omitted.
>
> I guess this has changed then. Or maybe I had just misinterpreted it.
> This is very good to know and is more restrictive than I had thought.
>
> > There is
> > currently a circumstance whereby conformance checking tools may suppress
> > errors relating to missing alt , but that does not mean the document is
> not
> > invalid due to its absence.
>
> It really is a double-edged sword. If you flag them as errors, it may
> promote accessibility. Or it may simply encourage authors to add null
> or bogus alt attribute values (which suggest an image alternative
> where none is present) simply to generate valid HTML. Admittedly, this
> is the current situation for HTML <5, so I wouldn't really expect
> things to change much.
>
> > I provide
> > detail and examples of this and a bridging technique (until
> > figure/figcaption is better supported) in section 3.12
>
> To me, it seems that it would be better to omit the alt attribute and
> <figcaption> altogether. In the case of unknown alternative text, any
> approach that utilizes alt or figcaption suggests that a text
> alternative is present, when in fact they do not contain an
> alternative at all but instead contain a placeholder for that
> alternative. This approach uses the alt attribute for something other
> than an alternative, which seems wrong to me. It suggests, "alt and/or
> figcaption present the image alternative text, except for when they
> don't, so good luck figuring out if they actually present an
> alternative or not."
>
> Allowing the absence of the alt attribute altogether could be
> absolutely clear - the image is inaccessible and does not have a known
> alternative. There's no ambiguity or burden on the user to
> differentiate between true alternative text and something else that is
> currently in the same place the alternative would go if it were known.
> Knowing definitively that something is inaccessible can be quite
> useful. AT could try to do something useful with the image.
> Additionally, automated accessibility validation tools could easily
> flag such images as inaccessible. If the alt and/or figcaption
> technique you provided was implemented when the alternative is not
> known, there would be no way to automatically differentiate them from
> other accessible images.
>
> Please clarify if I am still misunderstanding this.
>
> Jared
> > > >



--
with regards

Steve Faulkner
Technical Director - TPG

www.paciellogroup.com | www.HTML5accessibility.com |
www.twitter.com/stevefaulkner
HTML5: Techniques for providing useful text alternatives -
dev.w3.org/html5/alt-techniques/
Web Accessibility Toolbar - www.paciellogroup.com/resources/wat-ie-about.html